Tune DreamWaQ gait rewards

motrix_envs/src/motrix_envs/locomotion/go1/dreamwaq.py

@@ -124,8 +124,8 @@ class DreamWaQCfg(Go1WalkNpEnvCfg):
             "ang_vel_xy": -0.10,   # 抑制晃动
             "orientation": -0.5,   # 强制平稳姿态
             "dof_acc": -2.5e-7,
-            "base_height": -10.0,  # 强惩罚身高偏差，避免蹲伏或踮脚
-            "feet_air_time": 0.1,
+            "base_height": -5.0,
+            "feet_air_time": 0.03,
             "action_rate": -0.01,
             "joint_power": -2e-5,
             "smoothness": -0.02,
@@ -768,7 +768,7 @@ class DreamWaQTask(Go1WalkTask):
         air_time = info.get("air_time_at_contact")
         if first_contact is None or air_time is None:
             return np.zeros(self._num_envs, dtype=np.float32)
-        rew = np.sum((air_time - 0.5) * first_contact, axis=1)
+        rew = np.sum(np.maximum(air_time - 0.25, 0.0) * first_contact, axis=1)
         rew *= np.linalg.norm(commands[:, :2], axis=1) > 0.1
         return rew